search

aiidateam / aiida-common-workflows

A repository for the implementation of common workflow interfaces across materials-science codes and plugins
https://aiida-common-workflows.readthedocs.io
MIT License
52 stars 31 forks source link

WIP: Protocol modifications for VASP to comply with requirements of ACWF study #288

Closed espenfl closed 1 year ago

espenfl commented 2 years ago

Here we add necessary protocol updates for the upcoming study on unaries and oxides,

espenfl commented 2 years ago

@bosonie I have a question related to the fixation of the protocol. Should this be fixated to reproduce the results in the study, or to the closest analogy that would be used in practice? In particular I am thinking of the choice of the potentials for the lanthanides which we had to force to include all electrons in the valence to comply closest to the all-electron result. But in practice these potentials might not be used. Some feedback on this would be great as this also has a consequence of how we name the potential mapping.

bosonie commented 2 years ago

Hi @espenfl, the protocol must be the exact same you used to produce the results, since we want the results to be completely reproducible. Since you have more than one set, you need one protocol per set. For the moment it is up to you how to name them, then we will maybe agree on something together.

espenfl commented 2 years ago

@bosonie Okey, then we will leave the current one as is. We will update for the alternative if we want to include more than one set from VASP in the manuscript. Also, we will make some notes regarding this in the supplementary.

bosonie commented 2 years ago

Thanks @espenfl. There are few addition I will ask you to do.

First of all, since many codes do not want to identify the protocol used for this study as the "precise" protocol, we decided to have a separate protocol called "verification-PBE-v1". So please make a copy your "precise" protocol and put it under the name "verification-PBE-v1". We foresee in the future that the "precise" protocol will change, the "verification-PBE-v1" no.

Moreover, if possible, please write a function to extract the TS contribution from a VaspCommonRelaxWorkChain, look at this template: https://github.com/aiidateam/aiida-common-workflows/blob/master/aiida_common_workflows/workflows/relax/siesta/extractors.py

espenfl commented 2 years ago

@bosonie Yes, this sounds great. Having a protocol that is linked directly to the study makes more sense. Will update according to the requests.

bosonie commented 2 years ago

@espenfl will you have time to correct the small final details that are still mising?

espenfl commented 2 years ago

@bosonie Will do in the next few days.

espenfl commented 2 years ago

@bosonie Do we want the TS contribution in a separate module (as the example above) so that you can do post extraction or should we just add it as a calcfunction and store it in convert_outputs?

espenfl commented 1 year ago

Closing. Opening new PR for final.